HtmlToOpenXml使用中的特殊a标签问题

在使用HtmlToOpenXml库将html转换成word的时候有一些特殊标签是没有解析的。

比如我的html源中使用的锚链接是<a filepos=000456>link</a>，而目标是<a id="filepos456"></a>

该库将所有的a解析标签都写在HtmlConverterProcessTag.cs的ProcessLink方法里

 

代码如下

#region ProcessLink

		private void ProcessLink(HtmlEnumerator en)
		{
			String att = en.Attributes["href"];
			Hyperlink h = null;
			Uri uri = null;


			if (!String.IsNullOrEmpty(att))
			{
				// is it an anchor?
				if (att[0] == '#' && att.Length > 1)
				{
					// Always accept _top anchor
					if (!this.ExcludeLinkAnchor || att == "#_top")
					{
						h = new Hyperlink(
							) { History = true, Anchor = att.Substring(1) };
					}
				}
				// ensure the links does not start with javascript:
				else if (Uri.TryCreate(att, UriKind.Absolute, out uri) && uri.Scheme != "javascript")
				{
					HyperlinkRelationship extLink = mainPart.AddHyperlinkRelationship(uri, true);

					h = new Hyperlink(
						) { History = true, Id = extLink.Id };
				}
			}

			if (h == null)
			{
				// link to a broken url, simply process the content of the tag
				ProcessHtmlChunks(en, "</a>");
				return;
			}

			AlternateProcessHtmlChunks(en, "</a>");

			if (elements.Count > 0)
			{
				// Let's see whether the link tag include an image inside its body.
				// If so, the Hyperlink OpenXmlElement is lost and we'll keep only the images
				// and applied a HyperlinkOnClick attribute.
				List<OpenXmlElement> imageInLink = elements.FindAll(e => { return e.HasChild<Drawing>(); });
				if (imageInLink.Count != 0)
				{
					for (int i = 0; i < imageInLink.Count; i++)
					{
						// Retrieves the "alt" attribute of the image and apply it as the link's tooltip
						Drawing d = imageInLink[i].GetFirstChild<Drawing>();
						var enDp = d.Descendants<pic.NonVisualDrawingProperties>().GetEnumerator();
						String alt;
						if (enDp.MoveNext()) alt = enDp.Current.Description;
						else alt = null;

						d.InsertInDocProperties(
							 new a.HyperlinkOnClick() { Id = h.Id ?? h.Anchor, Tooltip = alt });
					}

					CompleteCurrentParagraph();
					AddParagraph(currentParagraph = htmlStyles.Paragraph.NewParagraph());
				}
				else
				{
					// Append the processed elements and put them to the Run of the Hyperlink
					h.Append(elements);

					if (!htmlStyles.DoesStyleExists("Hyperlink"))
					{
						htmlStyles.AddStyle("Hyperlink", new Style(
							new StyleName() { Val = "Hyperlink" },
							new UnhideWhenUsed(),
							new StyleRunProperties(
								new DocumentFormat.OpenXml.Wordprocessing.Color() { Val = "0000FF", ThemeColor = ThemeColorValues.Hyperlink },
								new Underline() { Val = UnderlineValues.Single }
							)
						) { Type = StyleValues.Character, StyleId = "Hyperlink" });
					}

					h.GetFirstChild<Run>().InsertInProperties(
						new RunStyle() { Val = htmlStyles.GetStyle("Hyperlink", StyleValues.Character) });

					this.elements.Clear();

					// Append the hyperlink
					elements.Add(h);
				}
			}
		}

		#endregion

实际上是只解析了href部分，那么对于特殊文件的特殊标签该怎么处理呢？在不改变大框架的前提下，只能在该方法内部添加我们需要的代码

由于我的链接是<a filepos=000456>link</a>和<a id="filepos456"></a>，所以我只修改了标签的定义

String att = en.Attributes["href"];
            if (att == null)
            {
                att = en.Attributes["filepos"];
                if (att != null)
                {
                    char[] c = { '0' };
                    att = "#filepos" + en.Attributes["filepos"].TrimStart(c); //锚点跟链接的不同是前面没有0，添加了filepos，所以对应修改
                }
                else
                {
                    att = en.Attributes["id"];
                }
            }

添加定义

BookmarkStart bs = null;
BookmarkEnd be = null;

赋值

if (att[0] == 'f' && att.Length > 1)
   {
   //添加书签
        bs = new BookmarkStart();
       bs.Name = att;
       bs.Id = att;
       be = new BookmarkEnd();
       be.Id = att;
 }

最后将节点添加到文档中即可

if (bs != null)
{
        elements.Add(bs);
        elements.Add(be);
}

转换成功后文档已正确添加书签

 

              祝贺该项目又解决掉一个bug！！！！！！！！